Qwen 3.6 27B benchmarked by community

Qwen 3.6 27B achieves high throughput with NVFP4 quantization on vLLM, with users reporting up to 100K context on 32GB VRAM at Q8. In a fantasy RP/agentic benchmark across 8 local models, Qwen 3.6 27B performed well relative to its parameter count.